Cumulus Support

Hello,
I searched a little bit in older Topics but found nothing what helped me.
I have a Fioneoffset WH3080 with Solar.
Currently (and in the past days) I have more then 8hours sunchine here in germany.

My solar values are looking similar like this:
Radiation 931 W/m2
UV Index 8.0
But when I look into the today/yesterday TAB I only have
Hours of Sunshine: 0.5 hrs

How are these sunchine hours calculated ....or better, whats going wrong ?

Jan

You need to compare the graph of theoretical solar value against measured value. Then adjust the theoretical value so the curve matches your measured values. I'm assuming you are using the default theoretical value calculation of Ryan-Stolzenbach.

In station setting, adjust the "RS Transmission factor", increase this to increase the theoretical value, and vice versa.

Once you have the curves matching pretty well - you want theoretical value to be slightly greater than measured value, and you need to do this on a cloudless day without haze - then whenever the measured value percentage of theoretical is greater than the settings "Sun Threshold (percent)" value, MX will start incrementing the sunshine minutes.

Here is an (old) example of my curves (blue=theoretical, red=measured)...

thank you Mark, for the quick answer .......I wil ltry....

You need to do the adjustments around local solar noon of course, so the values are at their maximum.

mcrossley wrote: ↑Tue 19 May 2020 1:34 pm you want theoretical value to be slightly greater than measured value

Is there a (simple) explanation for why you want it to be slightly greater around midday?

You want theoretical to be slightly greater because as the curve drops off in the morning and evening, the percentage value of a single W/m2 gets much greater. If the measured is just above theoretical you will find that it starts recording false sunshine in the morning and evening.

You will also find that the measured value can increase (considerably at times) if there are clouds around, they can reflect a lot of extra radiation down. So for broken cloud days that effect could lift the measured value above the sunshine threshold if your clear sky measured value is already above theoretical.

You could compensate for that by increasing the percent threshold, but I have found it is better to make measured value slightly lower.

There is also the "Solar Minimum (W/m2)" setting, which defines the minimum theoretical value below which it will not measure sunshine. This again is to stop false measurements at the low parts of the curve. The better your curves fit, the lower you can make that threshold value, but take care!

Mark, many thanks for that - I will revisit my fiddles of early last year when I was installing my Davis station

Today there was a bit of hazy high cloud around but still very sunny, you can see it dropped the percentage down to around 91-92% for most of the day. Those gin clear days are rare, though with the current lack of aircraft the skies are noticeably clearer than "normal" - not natural

Please let me know if this should be a new topic but I'm puzzled regarding hours of sunshine and log rollover.
Looking at the console Now Sunshine Today it did say 4,4 hrs around 09:50, Today/Yesterday Hours of Sunshine did say 4.4 and 13.7 for yesterday.
The Chart Sunshine Hours didn't show the 12th until after 10, as expected since I've Log Rollover 9am with Use 10am in Summer.
I can now see 13.7 for the 12th on Chart Sunshine Hours
I've added #SunshineHours to https://rogerthn.se/weather3/index.htm, it looks like this isn't meteorological day value?

Correct, sunshine hours is a midnight to midnight value.

mcrossley wrote: ↑Mon 13 Jul 2020 9:05 am Correct, sunshine hours is a midnight to midnight value.

Thanks!
indexT.htm updated

Getting there
I did increase Solar minimum (W/m2) from 0 to 10 i.e. sun hours is not increased if Solar Radiation is below 10?
Hours of sunshine yesterday did start to increase at 04:50, theoretical solar value was 0.
We'll see if I need to lower Solar minimum.

I think 10 is still very low - the error margins down at those levels are huge.

mcrossley wrote: ↑Wed 15 Jul 2020 9:21 am I think 10 is still very low - the error margins down at those levels are huge.

Thanks!
Trying with 15 and waiting for a morning without clouds, this morning it was 15 at 05:55 with a theoretical solar value of 39.